I have a fairly simple tank to tanks system where both the sustained and operating repsonse seems, well, flat wrong. I'm refering to the displacements. I've looked at the CII model and can't see any issues that would cause most of the nodes to move up under the sustained case.

Attached is the CII 5.3.2 system and config file.

Except the pressure.....

Regards,

And the effect will be more visible when you will turn on the Bourdon effect as you should do for FRP.

Regards,

Danb,



Otherwise, why are you being so cryptic? I'm not even looking at stresses yet, its the system response that puzzles me.

Ok, I didn't know that will be turned on automatically. FRP respect to steel is a sort of "spaghetti". Pressure will make pipes to elongate, this is why it will expand (even up). And I am cryptic because I still need to develop my comunication skill Many times I think that few words are enough.

Regards,

egcallis,

There is a vertical uniform load component +Y in the two load cases.

Did the loads get inputted to different axis / direction than intended?

I can understand the pipe expanding. But that's the outer shell. The CII node is at the centerline of the pipe. In any event, I'll do some more research but I still think that there's another issue.

Coade folks, is there a way to force the Bourdon effect off even when using Mat 20? I want to isolate my issue to the bourdon effect.

If you create a load case with only W (SUS) and if you will check again the displacements, you will understand. The only difference is the pressure.

Regards,

Well, I switched the Mat to A53B and made sure the Bourdon effect was set to none and the system is behaving like I expected. I then turned the Bourdon Effect on (translation Only) and got results similar (displacement wise) to the FRP Mat with Bourdon on. So Dan, you be right.

What I have done is placed spring cans under the two support points which really helps on nozzle loads, but the elbos are overstressing. Can the elbows be stiffened/reinforced or can I simply thicken the elbows or do I need to make the entire system thicker?

Richard, the uniform loads are seismic and are in Gs. But that's not the problem.

egcallis,

Sorry, I didn't look closely at the SUS load case components, no U1 U2.
Yes, you are right about the vertical displacements seen in SUS case.

If the vessel tower elements are changed to zero weight rigids, then the vertical displacements are reduced to 10% of the previous displacements. The vessel tower elements are growing that much from pressure.

Question - Do the vessel tower elements actually have some structured fill or packing saddles? The vessel tower elements might need some density to provide the weight of packing, so that seismic loads are more 'realistic'.

Okay, talked to the other stress engineer at my company and he pointed out that the larger shorter tank was also FRP so it to is also growing from the Bourdon effect. The other tanks are 316L, so the FRP tank is pusing up against the other two tanks.

In other news, I spoke with the client and they dropped the design pressure and temp. So with the addition of the spring cans, all is well.

Thanks Dan and Richard for the input.